Taking Aim at Wall Teichoic Acid Synthesis: New Biology and New Leads for Antibiotics

REVIEW ARTICLE

Taking aim at wall teichoic acid synthesis: new biology and new leads for antibiotics

Edward WC Sewell and Eric D Brown

Wall teichoic acids are a major and integral component of the Gram-positive cell wall. These structures are present across all species of Gram-positive bacteria and constitute roughly half of the cell wall. Despite decades of careful investigation, a deﬁnitive physiological function for wall teichoic acids remains elusive. Advances in the genetics and biochemistry of wall teichoic acid synthesis have led to a new understanding of the complexity of cell wall synthesis in Gram-positive bacteria. Indeed, these innovations have provided new molecular tools available to probe the synthesis and function of these cell wall structures. Among recent discoveries are unexpected roles for wall teichoic acid in cell division, coordination of peptidoglycan synthesis and b-lactam resistance in methicillin-resistant Staphylococcus aureus (MRSA). Notably, wall teichoic acid biogenesis has emerged as a bona ﬁde drug target in S. aureus, where remarkable synthetic-viable interactions among biosynthetic genes have been leveraged for the discovery and characterization of novel inhibitors of the pathway. The Journal of Antibiotics (2014) 67, 43–51; doi:10.1038/ja.2013.100; published online 30 October 2013

Keywords: b-lactam; chemical genetics; synthetic viability; wall teichoic acid

INTRODUCTION carried out during the 1950s and 1960s on strains of Bacilli, Wall teichoic acids are a major component of the Gram-positive cell Lactobacilli and Staphylocci.14 Nucleotide-activated ribitol-phosphate wall, present in roughly equal proportions to peptidoglycan and glycerol-phosphate were first identified in lysates of Lactobacillus (Figure 1).1 Initially thought to have an ancillary role in bacterial arabinosus and, based on their similarity to Park’s nucleotide physiology, wall teichoic acids have emerged in recent years as a critical and precursors to other glycosyl-based macromolecules, were component of the cell wall of Gram-positive bacteria. Functional roles predicted to have a role in cell wall formation.15,16 Follow-up for wall teichoic acids include determination of cell shape in the inquiry into the biological function of activated glycosyl units led model Gram-positive bacterium Bacillus subtilis, and in Staphylococcus to the first identification of poly(ribitol-phosphate) in the wall of aureus, these structures serve important roles in host colonization, L. arabinosus.17,18 Early examination of the structure of teichoic acids coordination of peptidoglycan synthesis and resistance to b-lactam showed the common presence of polymers composed of either 1–3 antibiotics.2–8 Interestingly, wall teichoic acid biosynthetic proteins are linked units of glycerol-3-phosphate or 1–5 linked units of ribitol-5- encoded by genes that exhibit paradoxical gene dispensability patterns; phosphate decorated with hexose or N-acetylhexosamine residues, 19,20 deletions in early genes in the pathway result in viable but non- respectively, and D-alanine. This work was followed by the infectious organisms, and exhibit synthetic-viable interactions with demonstration that both nucleotide precursors and wall teichoic otherwise essential late-acting genes.2,7,9,10 Synthetic viability describes acids were present across a wide range of Gram-positive a genetic interaction in which the lethal phenotype associated with organisms.14,21 As technologies in sample preparation and analysis interruption of an essential gene is suppressed by simultaneous advanced, a common linkage unit that anchored wall teichoic acid disruption of a second (non-essential) gene. The discovery of such polymers to peptidoglycan was described, containing (glycerol- synthetic-viable interactions in wall teichoic acid synthesis has enabled phosphate)-N-acetylmannosamine-b(1–4)-N-acetylglucosamine, and the rational discovery and characterization of novel antibacterial covalently linked to the 60 hydroxyl of N-acetylmuramic acid compounds that block wall teichoic acid biogenesis in S. aureus. residues of peptidoglycan (Figure 2).22–24 Advancements in tools for Indeed, where methicillin-resistant S. aureus (MRSA) is a leading molecular biology and biochemistry enabled elucidation of the Gram-positive pathogen in an infectious disease crisis of global scale, genetic requirements for wall teichoic acid biosynthesis, and the wall teichoic acid appears to be an exciting new target for antibacterial indispensable phenotype of various wall teichoic acid biosynthetic chemical matter of unique chemical class and mechanism.11–13 genes suggested an essential role for these structures in B. subtilis.25–27 Early foundational work on the chemistry and biology of wall Most recently, the development of soluble substrate analogs for teichoic acids, named after the Greek word teiwoz (teichos; wall), was teichoic acid biosynthetic intermediates has allowed for in vitro

Michael G DeGroote Institute for Infectious Disease Research and the Department of Biochemistry and Biomedical Sciences, McMaster University, Hamilton, ON, Canada Correspondence: Dr ED Brown, Michael G DeGroote Institute for Infectious Disease Research and the Department of Biochemistry and Biomedical Sciences, McMaster University, 1280 Main Street West, Hamilton, ON L8S4K1, Canada. E-mail: [email protected] Received 25 June 2013; revised 4 September 2013; accepted 10 September 2013; published online 30 October 2013 New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 44

acid is synthesized by tag (teichoic acid glycerol) or tar (teichoic acid ribitol) gene products. Regardless of polymer structure, wall teichoic acid synthesis occurs intracellularly on an undecaprenyl-phosphate lipid carrier that anchors the nascent polymer to the cytoplasmic face of the cellular membrane (Figure 3). Synthesis of a common linkage unit is initiated by TagO/TarO catalyzed transfer of an N-acetylglucosamine-1-phosphate residue from UDP-N-acetylglucosamine to undecaprenyl-phosphate to form the teichoic acid intermediate lipid a36 (Table 1 describes a straightforward nomenclature for teichoic acid intermediates based on the cognate enzyme.37 Lipid a is the substrate for TagA, lipid b is the substrate for TagB, etc. For polymeric substrates, such as the TagF substrate lipid f.n, n denotes the number of repeating units in the polymer). Sequentially, TagA/TarA catalyzes the transfer of an N-acetylmannosamine residue to lipid a to generate Figure 1 The Gram-positive cell wall. lipid b.28,29 Linkage unit synthesis is finalized by TagB/TarB with the addition of an sn-glycerol-3-phosphate unit yielding lipid f.1.28,38 Following linkage unit synthesis, steps for the completion of poly(glycerol-phosphate) and poly(ribitol-phosphate) polymers are unique to the respective polymers. In poly(glycerol-phosphate)- containing organisms, roughly 40 glycerol phosphate units are polymerized directly on lipid f.1 via the TagF enzyme.39–41 Lipid- linked f.40 is modified with multiple a-glucose units via the TagE enzyme, and exported to the extracellular surface of the cytoplasmic membrane via the TagGH ABC transport system.42–45 In contrast, in poly(ribitol-phosphate)-containing organisms, lipid f.1 is elaborated with an additional glycerol phosphate unit via the TarF enzyme to yield lipid f.2, before ribitol phosphate polymerization via TarL.10,32,46 Intracellular decoration of lipid l.40 in S. aureus includes modifications with both a-linked and b-linked N-acetylglucosamine, catalyzed by TarM and TarS, respectively.8,47 Similar to the export of poly(glycerol-phosphate), poly(ribitol- phosphate) is transported to the extracellular surface of the cytoplasmic membrane by the TarGH ABC transport system. Once exported, the lipid-linked teichoic acid polymer is thought to be D-alanylated by gene products from the dltABCD operon and transferred from the undecaprenol lipid carrier to peptidoglycan.48 The final transfer of wall teichoic acid polymers from undecaprenol Figure 2 Structures of common wall teichoic acids. (a) Poly(ribitol- phosphate to muramic acid of peptidoglycan is thought to be carried phosphate) wall teichoic acid from S. aureus.(b) Poly(glycerol-phosphate) out by a redundant set of three enzymes, namely LytR, Cps2a, Psr in wall teichoic acid from B. subtilis 168. Ala, alanine; Glc, glucose; GlcNAc, B. subtilis 168 and Msr, SA0908 and SA2101 in S. aureus.49,50 The N-acetylglucosamine. latter biosynthetic assignments await in vitro biochemical confirmation.

biochemical characterization of recombinant wall teichoic acid WALL TEICHOIC ACID IS A VIRULENCE FACTOR REQUIRED biosynthetic proteins.28–32 Cumulatively, this work led to the FOR HOST COLONIZATION deﬁnition of both the structures and biosynthetic requirements of Wall teichoic acid is believed to have diverse functional roles canonical poly(glycerol-phosphate) and poly(ribitol-phosphate) wall including cation binding, osmotic tolerance, heat tolerance, regulation teichoic acids in B. subtilis 168 and S. aureus, respectively (Figure 2). of autolysins, phage-binding and cell-shape determination.2,51–54 Further, the more recent discovery of the essential role for wall SYNTHESIS OF WALL TEICHOIC ACID IN B. SUBTILIS 168 teichoic acid in coordinating mechanisms required for host infection AND S. AUREUS and drug resistance in S. aureus has bolstered a concerted effort to A number of Gram-positive bacteria are known to produce wall uncover mechanisms underlying this physiology, and to discover teichoic acid polymers having complex repeating units containing, for inhibitors of wall teichoic acid biosynthesis for development as example, mannitol, erythritol, glucose or N-acetylglucosamine; how- antimicrobial leads.5,6,55,56 ever, a large proportion produce either poly(ribitol-phosphate) or The importance of wall teichoic acid as a virulence factor was ﬁrst poly(glycerol-phosphate) as the major polymer.33–35 Much of our elaborated by the Peschel group.3 In this study, the authors used a knowledge of the genetics and biochemistry of wall teichoic acid DtarO strain of S. aureus in a cotton rat nasal colonization model. synthesis has been elaborated in recent years in the model Gram- Remarkably, after a 7-day incubation period, there was a complete positive bacterium B. subtilis 168 and the pathogen S. aureus,which, absence of bacterial colonization in rats inoculated with the DtarO respectively, produce poly(glycerol-phosphate) and poly(ribitol- strain, whereas all rats in the test group inoculated with the isogenic phosphate) wall teichoic acids. In these organisms, wall teichoic wild-type strain were colonized (n ¼ 15 per group). Furthermore, the

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 45

Figure 3 Wall teichoic acid synthesis in (a) S. aureus and (b) B. subtilis 168.

DtarO mutation caused a decrease in the capacity of S. aureus to both teichoic acid in the virulence of S. aureus, and helped to establish and disseminate infection in a rabbit endocarditis model.58 substantiate wall teichoic acid biosynthesis as an antibacterial drug Careful investigation of the mechanistic underpinnings of the target. contribution of wall teichoic acid to these phenotypes was pursued. In an assessment of the effect of the DtarO genotype on the capacity COMPLEX DISPENSABILITY PATTERNS GIVE RISE TO of S. aureus to adhere to blood constituents, it was discovered that the SYNTHETIC-VIABLE GENETIC INTERACTIONS AMONG WALL mutation led to a reduction in bacterial adhesion to endothelial cells TEICHOIC ACID BIOSYNTHETIC GENES in in vitro models.58 These data established a role for wall teichoic Before the aforementioned reports investigating virulence of a DtarO acids in colonization via host tissue adhesion, and indicated a S. aureus strain, wall teichoic acids were thought to be essential connection between deﬁciencies in in vivo bacterial proliferation cellular structures. This conclusion was based on essential phenotypes and wall teichoic acid. The association between defects in wall teichoic evident for various wall teichoic acid biosynthetic genes.25–27 The acid and bacterial proliferation in vivo has been substantiated by revelation by the Peschel group that the tarO gene was dispensable in reports of additional DtarO phenotypes in cell division and in S. aureus led the Brown laboratory to undertake a meticulous susceptibility to innate immune components, including anti- reassessment of the dispensability of wall teichoic acid biosynthetic microbial peptides, defencins and antimicrobial fatty acids.5,57,59,60 genes in B. subtilis 168 and S. aureus strain RN4220, chosen for its Collectively, this work established a strict requirement for wall genetic malleability.2,9 These investigations conﬁrmed and extended

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 46

Table 1 Nomenclature for lipid-linked wall teichoic acid intermediates

Lipid-linked Enzyme substrate Chemical composition

TagA/ Lipid a GlcNAc-P-P-und TarA TagB/ Lipid b ManNAc-b(1–4)-GlcNAc-P-P-und TarB TarF Lipid f.1 GroP-ManNAc-b(1–4)-GlcNAc-P-P-und

TagF Lipid f.n (GroP)n-ManNAc-b(1–4)-GlcNAc-P-P-und

TarL Lipid l.n (RboP)n-(GroP)2-ManNAc-b(1–4)-GlcNAc-P-P- und

Abbreviations: GlcNAc, N-acetylglucosamine; GroP, sn-glycerol-3-phosphate; ManNAc, N-acetylmannosamine; P, phosphate; RboP, ribitol-5-phosphate; und, undecaprenol. Intermediates are named according to the cognate enzyme that utilizes the molecule as a substrate. For oligomeric and polymeric substrates, the number of repeating units is represented by n.

previous studies revealing essential phenotypes both in B. subtilis 168 and S. aureus for genes encoding wall teichoic acid priming (tagB, tarB, tarF), polymerization (tagF), nucleotide-activated precursor synthesis (tagD, tarD, tarI, tarJ) and export (tarH). Paradoxically, these investigations also confirmed that viable and stable deletions could be generated in the first step in wall teichoic acid biosynthesis carried out by the tagO/tarO gene product. This paradox was resolved Figure 4 Synthetic-viable interactions between early- and late-acting wall by demonstrating synthetic-viable genetic interactions between early teichoic acid biosynthetic genes. Deletions in late steps in wall teichoic and late steps. Indeed, both in B. subtilis 168 and S. aureus, the lethal acid biosynthesis are non-viable, whereas deletions in early steps are viable phenotypes of all of the so-called ‘essential’ wall teichoic acid genes and can rescue the lethal phenotype of a late-step interruption. Depicted is could be suppressed by deletion of tagO/tarO. The tagA/tarA gene, a series of examples using the early gene tarO and late genes tarGH. encoding the second step in wall teichoic acid synthesis, was also (a) Wild-type organisms with intact early- and late-acting wall teichoic found to be dispensable and deletions therein could similarly suppress acid biosynthetic genes synthesize wall teichoic acid and are viable. (b) Interruption of late-steps, DtarGH shown, blocks wall teichoic acid lethal phenotypes associated with loss of the downstream genes.7 The synthesis and is non-viable. (c) Interruption of early steps, DtarO shown, presence of two poly(ribitol-phosphate) polymerase genes in S. aureus blocks wall teichoic acid synthesis is and viable. (d) Simultaneous has confounded definitive assignment of the dispensability of these interruption of early and late steps blocks wall teichioc acid synthesis and genes, with indications that either one or both of the tarK and tarL is viable. Additional synthetic-viable combinations include DtarO/DtagO in polymerase genes are required to support growth in a wild-type combination with any late-acting gene deletion (tarBDFGHIJL/tagBDFGH), or background.10,46 It is clear, however, that non-viable interruptions in DtarA/DtagA in combination with any late-acting gene deletion polymerization step(s) can be rescued by deletion of tarO or tarA, (tarBDFGHIJL/tagBDFGH). similarly to remaining late-acting steps in wall teichoic acid synthesis. Through these genetic experiments, teichoic acid biosynthetic genes late-acting genes give rise to compromised fitness, albeit of varying have been grouped into dispensable early-acting genes (tagO, tagA, severity, that can be rescued by simultaneous inactivation of an early- tarO, tarA), or conditionally essential late-acting genes (tagB, tagD, acting gene. Genetic differences between S. aureus strains RN4220 and tagF, tarB, tarD, tarF, tarI, tarJ, tarK, tarL). Furthermore, late-acting COL may explain the varying severity of phenotypes associated with genes are essential in a wild-type background, but become dispensable deletions in late-acting steps within these strains. S. aureus strain in a DtagO/DtarO or DtagA/DtarA genetic background (Figure 4). RN4220 has been subjected to repeated mutagenesis to render it Interestingly, additional synthetic-viable interactions have been amenable to transformation with foreign DNA.61 This strain harbors noted recently in the MRSA strain COL. The TarG-mediated wall interruptions in sB regulatory machinery, preventing the expression teichoic acid export step in this strain is essential in a wild-type of associated stress response mechanisms.62 Stress sensing and background, however, viable mutations in strains resistant to inhibi- response machinery is a vital component of an organism’s ability to tors of the TarG exporter have been uncovered in the late-acting genes survive cellular stressors. In the absence of this system, interruption of tarB and tarD.55 Although tarB and tarD have essential phenotypes in late-acting steps in wall teichoic acid synthesis may go unrecognized S. aureus strain RN4220,9 characterization of the S. aureus COL tarB in strain RN4220 until a cell-death cascade is triggered. Conversely, in and tarD suppressor mutants suggested loss-of-function mutations, strain COL, a sB-mediated response could coordinate the required and teichoic acid was found to be absent from the cell walls of these machinery to mitigate the deleterious effects of inhibition of wall mutants. Nevertheless, tarB and tarD mutations conferred sensitivity teichoic acid biosynthesis. Ultimately, it is clear that genetic to osmotic, temperature and antibiotic stresses that could be rescued interactions among wall teichoic acid biosynthetic genes are by inactivation of tarO.55 Thus, although the observation of viable complex and strain-dependent. Nevertheless, strong trends have interruptions in late-acting genes in S. aureus COL is at odds with the emerged from a large body of work on this topic: the presence of binary essential/dispensable phenotypes documented in strain teichoic acid polymers in the cell walls of B. subtilis 168 or S. aureus RN4220, a general relationship remains consistent: interruptions in are not strictly essential, but deletion of a late-acting biosynthetic gene

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 47 results in a ﬁtness defect, the severity of which may range from Table 2 Wall teichoic acid synthesis inhibitors sensitization to antibiotic stress, to death, and is dependent on both the bacterial strain and identity of the interrupted gene. Furthermore, Chemical name or series Structure Target Reference late-acting genes display synthetic-viable interactions with early-acting Targocil TarG 56,65 genes, where the lethal phenotype caused by deletion of a late step can be suppressed by inactivation of an early step. N O Insight into the mechanism underlying synthetic-viable interactions N between early- and late-acting genes in wall teichoic acid synthesis was O O O N S gained through an unbiased systems-scale investigation charting N N gene–gene and gene–chemical interactions in B. subtilis 168.63 In this study, the promoter PywaC was identiﬁed through microarray Cl analysis as mounting an especially strong response following depletion of late enzymes in B. subtilis 168. Probing the PywaC promoter with a Tricyclicindoleacid TarG 55 well-characterized library of bioactive small molecules (bioactives) F O demonstrated a similar induction of the promoter by cell–wall N Cl OH bioactives, with a strong bias toward chemicals that inhibit undeca- Br prenol-linked metabolism. Furthermore, bioinformatic analysis exam- Cl ining gene–gene connectivity through genomic context inferred strong connectivity among genes involved in the biosynthesis of wall N-aryl triazole TarG 55 teichoic acid, undecaprenol and peptidoglycan. Taken together, these F F data suggested a model for the synthetic viability observed among early and late wall teichoic acid genes: deletion of a late-acting gene N N interrupts wall teichoic acid biosynthesis, resulting in the accumula- Cl N N tion of dead-end undecaprenol-linked intermediates. Accumulation of N undecaprenol-linked wall teichoic acid intermediates may deplete a limited cellular store of undecaprenol, similarly required for the C-aryl triazole TarG 55 synthesis of peptidoglycan, and thus result in cell death. Indeed, in a N N

B. subtilis 168 strain with the tagF polymerase gene under the control N of an inducible promoter, depletion of TagF resulted in a decrease 14 in incorporation of D-[ C]glutamate into newly synthesized peptidoglycan.63 Alternately, as the structural integrity of membrane bilayers is determined by the physicochemical properties of bilayer Ticlopidine TarO 6 N components, it is possible that accumulation of undecaprenol-linked wall teichoic acid intermediates on the inner leaflet of the cytoplasmic S Cl membrane has a direct destabilizing effect on the bilayer.64 In either scenario, the deleterious accumulation of undecaprenol-linked wall teichoic acid intermediates can be avoided by preventing metabolite flux into the pathway with the deletion of the tagO or tagA genes. subsequently used to rule out TarBDFIJL enzymes as potential targets. This pointed to either export or teichoic acid-peptidoglycan ligase SYNTHETIC VIABILITY OFFERS A PLATFORM FOR PATHWAY- enzymes as likely targets of targocil. Overexpression of tarGH was SPECIFIC, CELL-BASED SCREENING FOR WALL TEICHOIC found to confer resistance, whereas expression of tagGH from the ACID INHIBITORS unsusceptible B. subtilis 168 in S. aureus conferred complete The discovery of unique synthetic-viable genetic interactions among resistance. Mapping of spontaneous suppressor mutations wall teichoic acid genes underpinned the development of pathway- conferring resistance to the action of targocil also identified TarG, specific, cell-based screening for inhibitors of wall teichoic acid the translocase component of the ABC export complex, was the synthesis. The approach, which enables facile mechanistic follow-up antibiotic target. on whole-cell-bioactive compounds while eliminating off-target Similarly, Merck and Company used this methodology to uncover nuisance compounds, has been successfully used in both academe inhibitors of wall teichoic acid biosynthesis within its whole-cell- and the pharmaceutical sector to discover inhibitors of the wall bioactive collection of 20 000 anti-Staphylococcal small molecules.55 In teichoic acid transport protein TarG.55,56 First published by the this work, three diverse chemical classes of late-stage wall teichoic acid Walker laboratory, compounds that are growth inhibitory to inhibitors were uncovered: tricyclic indole acids, N-aryl-triazoles and S. aureus are counter-screened against a tarO deletion strain.56 C-aryl-triazoles (Table 2). Mapping of suppressor mutations to the Chemicals leading to growth inhibition in the wild-type, and not tarG gene once again implicated TarG as a target, and the inhibitors the DtarO strain, are predominantly late-step inhibitors of wall were found to be effective in reducing bacterial burden in a mouse teichoic acid synthesis. Compound 1835F03 was discovered and its thigh infection model. In both studies, the frequency of resistance to potency was optimized through a structure–activity relationship TarG inhibitors was high (B7 Â 106 at 8 Â MIC), as resistance could analysis to yield targocil (Table 2).65 Interestingly, targocil has be generated not only from mutations in the target protein but selective efficacy against S. aureus and no growth inhibitory effects additionally from loss-of-function mutations in tarO or tarA. on B. subtilis 168. In vitro enzymatic assays have been published for all However, as deletions in early steps in wall teichoic acid synthesis intracellular steps in wall teichoic acid synthesis and these assays were render S. aureus avirulent, TarG inhibitor-resistant mutants with

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 48

loss-of-function mutations in tarO or tarA were not viable in an phenotype to the b-O-N-acetylglucosamine modification, rather than in vivo model. to the presence of the teichoic acid polymer itself.8 This observation is Taken together, efforts to search for late-step inhibitors at Merck fascinating in the context of S. aureus wall teichoic acid modifications and in the Walker laboratory, respectively, suggest that TarG is a including both a-andb-O-N-acetylglucosamine, with only the highly druggable target that can be perturbed by chemicals of diverse b-linked sugar having a role in b-lactam resistance.8,47 structural class. The ease of discovery of TarG inhibitors is curious and due perhaps to its partial extracellular structure, or to its role as a TICLOPIDINE REVERSES b-LACTAM RESISTANCE IN MRSA: wall teichoic acid translocase protein. Furthermore, the challenging A NEW PROBE OF WALL TEICHOIC ACID SYNTHESIS AND A biochemistry of this multitransmembrane domain protein remains an PROMISING DRUG COMBINATION obstacle to rational in vitro optimization of bioactives targeting this Given the requirement for wall teichoic acid in b-lactam resistance, structure, and efforts continue to discover leads that are active against there is strong therapeutic potential for a small-molecule inhibitor of more tractable late-step targets in the wall teichoic acid biosynthetic an early step in teichoic acid synthesis. Toward this goal, the Brown pathway. laboratory developed a chemical combination screening strategy to exploit signature chemical–genetic interactions observed between A ROLE FOR WALL TEICHOIC ACID IN b-LACTAM RESISTANCE tarO and b-lactam antibiotics.6 A pairwise screen of 2080 IN MRSA previously approved drugs in combination with cefuroxime was The b-lactams are the most successful class of antibiotics in clinical conducted against MRSA strain USA300. Previously approved drugs history.66 Despite the clinical success of these molecules and the represent a rich source of bioactivity with the potential to be transformative role they have played in modern medicine, resistance repurposed for new therapeutic benefit.70,71 Furthermore, down- to all b-lactam antibiotics has been documented, often shortly after stream drug development efforts with repurposed bioactives are introduction into the clinic. Resistance to b-lactams is commonly advantaged by the availability of extensive safety and achieved via b-lactamase enzymes that inactivate the pharmacophore pharmacological data. While a screen for b-lactam sensitization by hydrolyzing its b-lactam ring, rendering the antibiotic inert.67 should enrich for compounds targeting early steps in wall teichoic To salvage efficacy of these drugs, b-lactams are commonly acid synthesis, Tan et al.72 recently demonstrated that inhibition of coformulated with a b-lactamase inhibitor, (e.g. Augmentin; other cellular targets in S. aureus, including peptidoglycan amoxicillin þ clavulanic acid). Although this approach is broadly biosynthesis and cell division, could likewise sensitize MRSA to successful for the treatment of Gram-negative infections, clinically b-lactams. Accordingly, molecules discovered to restore the efficacy important Gram-positive pathogens such as MRSA achieve b-lactam- of cefuroxime against MRSA were advanced to a counter screen to resistance via additional mechanisms. In MRSA, resistance is identify those that caused b-lactam sensitization via inhibition of an determined in part by the presence of a horizontally acquired early step in wall teichoic acid biosynthesis. Bioactive molecules penicillin-binding protein (PBP), PBP2A, which is unsusceptible to showing synergy in the primary screen were tested in a DtarO inactivation by b-lactam antibiotics.68 While the presence of PBP2A background for suppression of cefuroxime sensitization. Ticlopidine was once thought to be the sole determinant for b-lactam resistance (Ticlid), an antihypertension drug that interrupts ADP-dependent in S. aureus, it has become increasingly clear that this phenotype is platelet formation, was the only molecule to advance from this supported by a complex network of resistance determinants, counter screen.73 Ticlopidine also showed signature synthetic-viable including PBPs (PBP2 and PBP4), and intriguingly, wall teichoic acid. interactions with late-acting wall teichoic acid biosynthetic genes, In 1994, the Murakami group69 conducted transposon-mediated suppressing the lethal phenotype caused by genetic or chemical mutagenesis of a clinical MRSA isolate to identify methicillin inactivation of TarG. While ticopidine did not display any growth- resistance determinants. The authors demonstrated that selective inhibition activity, the ticlopidine–cefuroxime combination displayed disruption of the llm gene could reverse high-level methicillin potent and synergistic in vitro efficacy against 10 clinical S. aureus resistance in clinical isolates of MRSA. The llm gene was identified isolates, including nine MRSA strains, and efficacy in an in vivo as the wall teichoic acid biosynthetic gene tarO almost a decade later S. aureus infection model. Phenotypes for ticlopine-treated S. aureus, and a focused evaluation of the connectivity between wall teichoic including sensitization to b-lactam antibiotics, absence of teichoic acid biosynthesis and b-lactam resistance in MRSA ensued.5,36 This acid from the cell wall and resistance to teichoic acid-specific phage, work demonstrated that selective genetic inactivation of tarO caused a were consistent with phenotypes for tarO/tarA gene deletions, and b-lactam sensitization phenotype in both methicillin-sensitive and provided further support for a mechanism for ticlopidine in inhibi- methicillin-resistant strains of S. aureus.Theb-lactam sensitization tion of an early-step in wall teichoic acid biosynthesis. In vitro enzyme phenotype was observed only upon interruption of the early-acting assays confirmed TarO to be the target, and the ticlopidine– enzyme TarO and not the late-acting enzyme TarG, although cefuroxime combination was found to be selective for S. aureus. coadministration of imipenem or oxacillin with a TarG inhibitor The availability of a small-molecule inhibitor for TarO provided an was able to elicit a three-log decrease in the frequency of TarG opportunity to probe the mechanism of synergy between early steps inhibitor-resistant mutations.5,55 Furthermore, the b-lactam in wall teichoic acid biosynthesis and transpeptidation steps in sensitization phenotype was found to be heterogeneous across a peptidoglycan biosynthesis. Indeed, well-characterized bioactive small spectrum of b-lactam antibiotics, suggesting a very specific interaction molecules have frequently proved highly useful in understanding between wall teichoic acids and peptidoglycan biosynthetic complex biological processes.74–76 Here, growth-inhibition pheno- machinery. Further investigation of the connection between wall types in MRSA were explored with a diverse collection b-lactam teichoic acid and peptidoglycan synthesis by electron microscopy antibiotics in conjunction with TarO inhibition. Interestingly, uncovered clear phenotypes for defects in wall teichoic acid synthesis blocking TarO function either with ticlopidine or by gene deletion in cell division in S. aureus; interruption of TarO was linked to defects sensitized MRSA only to b-lactams with a high affinity for PBP2, in cell septation and cell separation.5 Further analysis of the structural particularly the cephalosporins, and not to all b-lactams.6 The requirements for wall teichoic acid in b-lactam resistance linked this selective interaction of ticlopidine with PBP2-specific b-lactams is

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 49

Figure 5 A model for wall teichoic acid in b-lactam resistance in MRSA. (a) In the absence of b-lactam challenge, transglycosylase (TG) and transpeptidase (TP) domains in PBP2 work in concert with PBP2A to produce high levels of primary crosslinked peptidoglycan (PG). Wall teichoic acid (WTA) recruits PBP4 to the division septum where it catalyzes formation of highly crosslinked PG. (b) Under b-lactam challenge, the TP site of PBP2 is inhibited. TG activity of PBP2 and TP activity of PBP2A work in concert to produce reduced levels of primary crosslinked PG compared with the unchallenged condition. Primary crosslinked PG is further crosslinked by PBP4 in a WTA-dependent manner. (c) Under simultaneous challenge with a b-lactam and WTA inhibitor, PBP4 is mislocalized and does not synthesize highly crosslinked PG. The presence of low levels of primary crosslinked PG synthesized by PBP2 and PBP2A under b-lactam challenge are insufficient to support life. fascinating in the context of an emerging literature on the unique between the TarO inhibitor ticlopidine and PBP2-specific antibiotics: functions of individual PBPs in MRSA. PBP2 is the only bifunctional PBP2-specific b-lactams inhibit the transpeptidation activity of PBP2, PBP with transpeptidase and transglycosylase activities in S. aureus resulting in lowered levels of primary crosslinked cell wall, whereas and is essential to support life.77,78 Localization of PBP2 to the ticlopidine inhibits the synthesis of wall teichoic acid, preventing division septum is required for the essential transglycosylase activity proper localization and function of PBP4, and preventing secondary of this enzyme, and is mediated by an interaction between the cell wall crosslinking. The low level of primary crosslinked cell wall transpeptidase active site and its transpeptidation substrate.79 Under synthesized under this challenge is insufficient to support life b-lactam challenge, the transpeptidase active site on PBP2 is acylated, (Figure 5). Although untested at this time, it appears likely that and autonomous localization to the division septum is inhibited. b-lactam sensitivity reported to be associated with the loss of b-O-N- Under this stress, PBP2A, a horizontally acquired transpeptidase acetylglucosamine modifications to wall teichoic acid may also be due unsusceptible to inhibition by b-lactam antibiotics, carries out to a loss of PBP4 localization.8 Indeed, it is an intriguing idea that the transpeptidation activities, and acts as a structural scaffold to wall teichoic acid polymer may well be a specific scaffold for subtle recruit acylated PBP2 to the division septum.77,79 In this manner, functions in peptidoglycan synthesis and cell division that are PBP2 remains properly localized under b-lactam challenge and can nevertheless absolutely critical to the complex manifestation of perform its essential transglycosylation function. This cooperative b-lactam resistance in MRSA. function of PBP2 and PBP2A yields peptidoglycan strands with low- level crosslinking, up to approximately five units in length.4,80 Glycan CONCLUDING REMARKS strands with low-level crosslinking are substrates for PBP4, which Wall teichoic acid biosynthesis has recently emerged as an exciting functions as a secondary transpeptidase at the division septum to new target in antibiotic research. Long thought to be ancillary generate highly crosslinked peptidoglycan.4,81 Thus, the activities of structures in the cell wall of Gram-positive organisms, new roles for PBP2, PBP2A and PBP4 are required to synthesize the highly wall teichoic acid in the coordination of cell division, peptidoglycan crosslinked cell wall in MRSA. Indeed, many MRSA strains are able synthesis and resistance to b-lactam antibiotics have emerged. A to survive inhibition of transpeptidase activities on either PBP2 or detailed understanding of the biology underpinning these events and PBP4, albeit with decreased amounts of highly crosslinked cell wall. the description of unique chemical and genetic interactions within However, simultaneous inhibition of both PBP2 and PBP4 these processes have provided the opportunity to develop focused transpeptidase activities is lethal.82 Remarkably, Atilano and co- cell-based screening tools for the discovery of inhibitors of both wall workers demonstrated that the localization of PBP4 at the division teichoic acid biosynthesis and b-lactam resistance in MRSA. Indeed, site is dependent on the presence of the wall teichoic acid polymer.4 mechanistic follow-up on bioactives from high-throughput screening Taken together, these data allowed the Brown group to propose a efforts is a time- and resource-intensive endeavor with low rates of model for the mechanistic basis for the synergy observed in MRSA success. The pathway-specific, cell-based screening strategies

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 50

highlighted herein have exploited synthetic-viable interactions to 22 Kaya, S., Yokoyama, K., Araki, Y. & Ito, E. N-acetylmannosaminyl(1–4)N-acetylgluco- enable focused discovery of pathway-specific bioactives, facilitating samine, a linkage unit between glycerol teichoic acid and peptidoglycan in cell walls of several Bacillus strains. J. Bacteriol. 158, 990–996 (1984). streamlined follow-up, elimination of nuisance compounds and 23 Kojima, N., Araki, Y. & Ito, E. Structure of the linkage units between ribitol teichoic ultimately, the discovery of inhibitors for TarO and TarG. Indeed, acids and peptidoglycan. J. Bacteriol. 161, 299–306 (1985). TarG appears to be a highly druggable target that is susceptible to a 24 Coley, J., Tarelli, E., Archibald, A. R. & Baddiley, J. The linkage between teichoic acid and peptidoglycan in bacterial cell walls. FEBS Lett. 88, 1–9 (1978). variety of chemical classes. Ticlopidine, a selective inhibitor of TarO, is 25 Maue¨l, C., Young, M., Margot, P. & Karamata, D. The essential nature of teichoic acids strongly synergistic with PBP2-specific b-lactam antibiotics and in Bacillus subtilis as revealed by insertional mutagenesis. Mol. Gen. Genet. 215, 388–394 (1989). restores the efficacy of these once-successful drugs for treatment of 26 Bhavsar, A. P., Beveridge, T. J. & Brown, E. D. Precise deletion of tagD and controlled MRSA infections. Cumulatively, inhibitors of wall teichoic acid depletion of its product, glycerol 3-phosphate cytidylyltransferase, leads to irregular synthesis offer an exciting opportunity for the development of novel morphology and lysis of Bacillus subtilis grown at physiological temperature. J. Bacteriol. 183, 6688–6693 (2001). antibacterial leads with a new mechanism of action to treat drug- 27 Bhavsar, A. P., Erdman, L. K., Schertzer, J. W. & Brown, E. D. Teichoic acid is an resistant Staphylococcal infections. essential polymer in Bacillus subtilis that is functionally distinct from teichuronic acid. J. Bacteriol. 186, 7865–7873 (2004). 28 Ginsberg, C., Zhang, Y.-H., Yuan, Y. & Walker, S. In vitro reconstitution of two essential steps in wall teichoic acid biosynthesis. ACS Chem. Biol. 1, 25–28 (2006). ACKNOWLEDGEMENTS 29 Zhang, Y.-H., Ginsberg, C., Yuan, Y. & Walker, S. Acceptor substrate selectivity and kinetic mechanism of Bacillus subtilis TagA. Biochemistry 45, 10895–10904 (2006). This work was supported by a Canada Research Chair salary award to EDB 30 Pereira, M. P. et al. The wall teichoic acid polymerase TagF efficiently synthesizes and operating funding from the Canadian Institutes of Health Research poly(glycerol phosphate) on the TagB product lipid III. Chem. Bio. Chem. 9, (MOP-15496). 1385–1390 (2008). 31 Sewell, E. W., Pereira, M. P. & Brown, E. D. The wall teichoic acid polymerase TagF is non-processive in vitro and amenable to study using steady state kinetic analysis. J. Biol. Chem. 284, 21132–21138 (2009). 32 Brown, S., Zhang, Y.-H. & Walker, S. A revised pathway proposed for Staphylococcus 1 Hancock, I. C. Bacterial cell surface carbohydrates: structure and assembly. Biochem. aureus wall teichoic acid biosynthesis based on in vitro reconstitution of the Soc. Trans. 24, 183–187 (1997). intracellular steps. Chem. Biol. 15, 12–21 (2008). 2 D’Elia, M. A., Millar, K. E., Beveridge, T. J. & Brown, E. D. Wall teichoic acid polymers 33 Archibald, A. R. in Advances in Microbial Physiology Vol. 11 (eds Rose, A. H.) 53–96 are dispensable for cell viability in Bacillus subtilis. J. Bacteriol. 188, 8313–8316 (Academic Press, London, 1974). (2006). 34 Naumova, I. B. et al. Cell wall teichoic acids: structural diversity, species specificity in 3Weidenmaier,C.et al. Role of teichoic acids in Staphylococcus aureus nasal the genus Nocardiopsis, and chemotaxonomic perspective. FEMS Microbiol. Rev. 25, colonization, a major risk factor in nosocomial infections. Nat. Med. 10, 243–245 269–284 (2001). (2004). 35 Hancock, I. C. & Baddiley, J. Biosynthesis of the wall teichoic acid in Bacillus 4 Atilano, M. L. et al. Teichoic acids are temporal and spatial regulators of peptidoglycan licheniformis. Biochem. J. 127, 27–37 (1972). cross-linking in Staphylococcus aureus. Proc. Natl Acad. Sci. USA 107, 36 Soldo, B., Lazarevic, V. & Karamata, D. TagO is involved in the synthesis of all anionic cell-wall polymers in Bacillus subtilis 168. Microbiology (Reading, UK) 148, 18991–18996 (2010). 5 Campbell, J. et al. Synthetic lethal compound combinations reveal a fundamental 2079–2087 (2002). 37 Pereira, M. P. & Brown, E. D. in Microbial Glycobiology: Structures, Relevance And connection between wall teichoic acid and peptidoglycan biosyntheses in Staphylo- Applications (eds Moran, A. P., Holst, O., Brennan, P. & Itzstein, von, M.) 337–348 coccus aureus. ACS Chem. Biol. 6, 106–116 (2011). (Academic Press, Burlington, MA, 2009). 6 Farha, M. A. et al. Inhibition of WTA synthesis blocks the cooperative action of PBPs 38 Bhavsar, A. P., Truant, R. & Brown, E. D. The TagB protein in Bacillus subtilis 168 is an and sensitizes MRSA to b-lactams. ACS Chem. Biol. 8, 226–233 (2013). intracellular peripheral membrane protein that can incorporate glycerol phosphate onto 7 D’Elia, M. A., Henderson, J. A., Beveridge, T. J., Heinrichs, D. E. & Brown, E. D. The a membrane-bound acceptor in vitro. J. Biol. Chem 280, 36691–36700 (2005). N-acetylmannosamine transferase catalyzes the first committed step of teichoic acid 39 Schertzer, J. W. & Brown, E. D. Purified, recombinant TagF protein from Bacillus assembly in Bacillus subtilis and Staphylococcus aureus. J. Bacteriol. 191, subtilis 168 catalyzes the polymerization of glycerol phosphate onto a membrane 4030–4034 (2009). acceptor in vitro. J. Biol. Chem. 278, 18002–18007 (2003). 8 Brown, S. et al. Methicillin resistance in Staphylococcus aureus requires glycosylated 40 Schertzer, J. W., Bhavsar, A. P. & Brown, E. D. Two conserved histidine residues wall teichoic acids. Proc. Natl Acad. Sci. USA 109, 18909–18914 (2012). are critical to the function of the TagF-like family of enzymes. J. Biol. Chem. 280, 9 D’Elia, M. A. et al. Lesions in teichoic acid biosynthesis in Staphylococcus aureus lead 36683–36690 (2005). to a lethal gain of function in the otherwise dispensable pathway. J. Bacteriol. 188, 41 Schertzer, J. W. & Brown, E. D. Use of CDP-glycerol as an alternate acceptor for the 4183–4189 (2006). teichoic acid polymerase reveals that membrane association regulates polymer length. 10 Pereira, M. P., D’Elia, M. A., Troczynska, J. & Brown, E. D. Duplication of teichoic acid J. Bacteriol. 190, 6940–6947 (2008). biosynthetic genes in Staphylococcus aureus leads to functionally redundant poly 42 Young, F. E., Smith, C. & Reilly, B. E. Chromosomal location of genes regulating (ribitol phosphate) polymerases. J. Bacteriol. 190, 5642–5649 (2008). resistance to bacteriophage in Bacillus subtilis. J. Bacteriol. 98, 1087–1097 (1969). 11 Klevens, R. M. et al. Invasive methicillin-resistant Staphylococcus aureus infections in 43 Honeyman, A. L. & Stewart, G. C. The nucleotide sequence of the rodC operon of the United States. JAMA 298, 1763–1771 (2007). Bacillus subtilis. Mol. Microbiol. 3, 1257–1268 (1989). 12 Boucher, H. W. et al. Bad bugs, no drugs: no ESKAPE! An Update from the Infectious 44 Allison, S. E., D’Elia, M. A., Arar, S., Monteiro, M. A. & Brown, E. D. Studies of the Diseases Society of America. Clin. Infect. Dis. 48, 1–12 (2009). genetics, function, and kinetic mechanism of TagE, the wall teichoic acid glycosyl- 13 Brown, E. D. Is the GAIN Act a turning point in new antibiotic discovery? Can. J. transferase in Bacillus subtilis 168. J. Biol. Chem. 286, 23708–23716 (2011). Microbiol. 59, 153–156 (2013). 45 Lazarevic, V. & Karamata, D. The tagGH operon of Bacillus subtilis 168 encodes a two- 14 Armstrong, J. J., Baddiley, J., Buchanan, J. G., Carss, B. & Greenberg, G. R. Isolation component ABC transporter involved in the metabolism of two wall teichoic acids. Mol. and structure of ribitol phosphate derivatives (teichoic acids) from bacterial cell walls. Microbiol. 16, 345–355 (1995). J. Chem. Soc. 4344–4354 (1958). 46 Meredith, T. C., Swoboda, J. G. & Walker, S. Late-stage polyribitol phosphate wall 15 Baddiley, J. & Mathias, A. P. Cytidine nucleotides. Part I. Isolation from Lactobacillus teichoic acid biosynthesis in Staphylococcus aureus. J. Bacteriol. 190, 3046–3056 arabinosus. J. Chem. Soc. 2723–2731 (1954). (2008). 16 Baddiley, J., Buchanan, J. G., Carss, B., Mathias, A. P. & Sanderson, A. R. The 47 Xia, G. et al. Glycosylation of wall teichoic acid in Staphylococcus aureus by TarM. isolation of cytidine diphosphate glycerol, cytidine diphosphate ribitol and J. Biol. Chem. 285, 13405–13415 (2010). mannitol 1-phosphate from Lactobacillus arabinosus. Biochem. J. 64, 599–603 48 Neuhaus, F. C. & Baddiley, J. A continuum of anionic charge: structures and functions (1956). of D-alanyl-teichoic acids in gram-positive bacteria. Microbiol. Mol. Biol. Rev. 67, 17 Baddiley, J., Buchanan, J. G. & Greenberg, G. R. A substance containing glyceropho- 686–723 (2003). sphate and ribitol phosphate residues in Lactobacillus arabinosus. Biochem. J. 66, 49 Kawai, Y. et al. A widespread family of bacterial cell wall assembly proteins. EMBO J. 51P–52P (1957). 30, 4931–4941 (2011). 18 Baddiley, J., Buchanan, J. G. & Carss, B. The presence of ribitol phosphate in bacterial 50 Dengler, V. et al. Deletion of hypothetical wall teichoic acid ligases in Staphylococcus cell walls. Biochim. Biophys. Acta 27, 220 (1958). aureus activates the cell wall stress response. FEMS Microbiol. Lett. 333, 109–120 19 Burger, M. M. & Glaser, L. The synthesis of teichoic acids I. Polyglycerophosphate. (2012). J. Biol. Chem. 239, 3168–3177 (1964). 51 Hoover, D. G. & Gray, R. J. Function of cell wall teichoic acid in thermally injured 20 Glaser, L. The synthesis of teichoic acids. II. Polyribitol phosphate. J. Biol. Chem. 239, Staphylococcus aureus. J. Bacteriol. 131, 477–485 (1977). 3178–3186 (1964). 52 Vergara-Irigaray, M. et al. Wall teichoic acids are dispensable for anchoring the PNAG 21 Clarke, P. H., Glover, P. & Mathias, A. P. The occurrence of polyol derivatives of cytidine exopolysaccharide to the Staphylococcus aureus cell surface. Microbiology 154, diphosphate in micro-organisms. J. Gen. Microbiol. 20, 156–164 (1959). 865–877 (2008).

The Journal of Antibiotics New biology and leads in teichoic acid synthesis EWC Sewell and ED Brown 51

53 Young, F. E. Requirement of glucosylated teichoic acid for adsorption of phage in 68 Lim, D. & Strynadka, N. C. J. Structural basis for the beta-lactam resistance of PBP2a Bacillus subtilis 168. Proc. Natl Acad. Sci. USA 58, 2377–2384 (1967). from methicillin-resistant Staphylococcus aureus. Nat. Struct. Biol. 9, 870–876 (2002). 54 Schlag, M. et al. Role of Staphylococcal wall teichoic acid in targeting the major 69 Maki, H., Yamaguchi, T. & Murakami, K. Cloning and characterization of a gene autolysin Atl. Mol. Microbiol. 75, 864–873 (2010). affecting the methicillin resistance level and the autolysis rate in Staphylococcus 55 Wang, H. et al. Discovery of wall teichoic acid inhibitors as potential anti-MRSA aureus. J. Bacteriol. 176, 4993–5000 (1994). b-lactam combination agents. Chem. Biol. 20, 272–284 (2013). 70 Ejim, L. et al. Combinations of antibiotics and nonantibiotic drugs enhance anti- 56 Swoboda, J. G. et al. Discovery of a small molecule that blocks wall teichoic acid microbial efficacy. Nat. Chem. Biol. 7, 348–350 (2011). biosynthesis in Staphylococcus aureus. ACS Chem. Biol. 4, 875–883 (2009). 71 Spitzer, M. et al. Cross-species discovery of syncretic drug combinations that 57 Peschel, A. et al. Inactivation of the dlt Operon in Staphylococcus aureus confers potentiate the antifungal fluconazole. Mol. Syst. Biol. 7, 499 (2011). sensitivity to defensins, protegrins, and other antimicrobial peptides. J. Biol. Chem. 72 Tan, C. M. et al. Restoring methicillin-resistant Staphylococcus aureus susceptibility to 274, 8405–8410 (1999). beta-lactam antibiotics. Sci. Transl. Med. 4, 126ra35 (2012). 58 Weidenmaier, C. et al. Lack of wall teichoic acids in Staphylococcus aureus leads to 73 O’Brien, J. R., Etherington, M. D. & Shuttleworth, R. O. Ticlopidine—an antiplatelet reduced interactions with endothelial cells and to attenuated virulence in a rabbit drug: effects in human volunteers. Thromb. Res. 13, 245–254 (1978). model of endocarditis. J. Infect. Dis. 191, 1771–1777 (2005). 74 Farha, M. A. & Brown, E. D. Chemical probes of Escherichia coli uncovered through 59 Kohler, T., Weidenmaier, C. & Peschel, A. Wall teichoic acid protects Staphylococcus chemical–chemical interaction profiling with compounds of known biological activity. aureus against antimicrobial fatty acids from human skin. J. Bacteriol. 191, Chem. Biol. 17, 852–862 (2010). 4482–4484 (2009). 75 Root, D. E., Flaherty, S. P., Kelley, B. P. & Stockwell, B. R. Biological mechanism 60 Weidenmaier, C. & Peschel, A. Teichoic acids and related cell-wall glycopolymers in profiling using an annotated compound library. Chem. Biol. 10, 881–892 (2003). Gram-positive physiology and host interactions. Nat. Rev. Microbiol. 6, 276–287 76 Falconer, S. B., Czarny, T. L. & Brown, E. D. Antibiotics as probes of biological (2008). complexity. Nat. Chem. Biol. 7, 415–423 (2011). 61 Nair, D. et al. Whole-genome sequencing of Staphylococcus aureus strain RN4220, a 77 Pinho, M. G., de Lencastre, H. & Tomasz, A. An acquired and a native penicillin- key laboratory strain used in virulence research, identifies mutations that affect not binding protein cooperate in building the cell wall of drug-resistant staphylococci. only virulence factors but also the fitness of the strain. J. Bacteriol. 193, 2332–2335 Proc. Natl Acad. Sci. USA 98, 10886–10891 (2001). (2011). 78 Pinho, M. G., Filipe, S. R., de Lencastre, H. & Tomasz, A. Complementation of the 62 van Schaik, W. & Abee, T. The role of sigmaB in the stress response of Gram-positive essential peptidoglycan transpeptidase function of penicillin-binding protein 2 (PBP2) bacteria - targets for food preservation and safety. Curr. Opin. Biotechnol. 16, by the drug resistance protein PBP2A in Staphylococcus aureus. J. Bacteriol. 183, 218–224 (2005). 6525–6531 (2001). 63 D’Elia, M. A. et al. Probing teichoic acid genetics with bioactive molecules reveals new 79 Pinho, M. G. & Errington, J. Recruitment of penicillin-binding protein PBP2 to the interactions among diverse processes in bacterial cell wall biogenesis. Chem. Biol. 16, division site of Staphylococcus aureus is dependent on its transpeptidation substrates. 548–556 (2009). Mol. Microbiol. 55, 799–807 (2005). 64 Zhang, Y.-M. & Rock, C. O. Membrane lipid homeostasis in bacteria. Nat. Rev. 80 Leski, T. A. & Tomasz, A. Role of penicillin-binding protein 2 (PBP2) in the antibiotic Microbiol. 6, 222–233 (2008). susceptibility and cell wall cross-linking of Staphylococcus aureus: evidence for the 65 Lee, K., Campbell, J., Swoboda, J. G., Cuny, G. D. & Walker, S. Development of cooperative functioning of PBP2, PBP4, and PBP2A. J. Bacteriol. 187, 1815–1824 improved inhibitors of wall teichoic acid biosynthesis with potent activity against (2005). Staphylococcus aureus. Bioorg. Med. Chem. Lett. 20, 1767–1770 (2010). 81 Wyke, A. W., Ward, J. B., Hayes, M. V. & Curtis, N. A. A role in vivo for penicillin- 66 Walsh, C. Antibiotics: Actions, Origins, Resistance 1–335 (ASM Press, Washington, binding protein-4 of Staphylococcus aureus. Eur. J. Biochem. 119, 389–393 (1981). DC, 2003). 82 Memmi, G., Filipe, S. R., Pinho, M. G., Fu, Z. & Cheung, A. Staphylococcus aureus 67 Drawz, S. M. & Bonomo, R. A. Three decades of beta-lactamase inhibitors. Clin. PBP4 is essential for lactam resistance in community-acquired methicillin-resistant Microbiol. Rev. 23, 160–201 (2010). strains. Antimicrob. Agents. Chemother. 52, 3955–3966 (2008).

The Journal of Antibiotics